¿Por qué diantres no funciona esto?

failed_transformer

from transformers import AutoTokenizer, AutoModelForQuestionAnswering, Trainer, 
                         TrainingArguments, default_data_collator
import datasets

model_name = 'dccuchile/bert-base-spanish-wwm-cased'

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForQuestionAnswering.from_pretrained(model_name)

train_data = datasets.load_dataset('squad_es', 'v1.1.0', split='train[:80%]')
eval_data = datasets.load_dataset('squad_es', 'v1.1.0', split='train[80%:]')

def preprocess_data(examples):
    return tokenizer(examples['question'], examples['context'], truncation=True, padding='max_length')

train_data = train_data.map(preprocess_data, batched=True)
eval_data = eval_data.map(preprocess_data, batched=True)

training_args = TrainingArguments(output_dir='./results',    
    evaluation_strategy = "epoch", learning_rate=2e-5,
    per_device_train_batch_size=16, per_device_eval_batch_size=64,
    num_train_epochs=3,    weight_decay=0.01)

trainer = Trainer(model=model, args=training_args,
    train_dataset=train_data, eval_dataset=eval_data,
    data_collator=default_data_collator,tokenizer=tokenizer)

trainer.train()

el error que me da es:
Traceback (most recent call last): File "/Users/sergio/Sandbox/chatGPT/torch-train.py", line 39, in <module> trainer.train()
File "/Users/sergio/Library/Python/3.9/lib/python/site-packages/transformers/trainer.py", line 1633, in train return inner_training_loop(
File "/Users/sergio/Library/Python/3.9/lib/python/site-packages/transformers/trainer.py", line 1902, in _inner_training_loop tr_loss_step = self.training_step(model, inputs)
File "/Users/sergio/Library/Python/3.9/lib/python/site-packages/transformers/trainer.py", line 2645, in training_step loss = self.compute_loss(model, inputs)
File "/Users/sergio/Library/Python/3.9/lib/python/site-packages/transformers/trainer.py", line 2690, in compute_loss raise ValueError( ValueError: The model did not return a loss from the inputs, only the following keys: start_logits,end_logits. For reference, the inputs it received are input_ids,token_type_ids,attention_mask.